Towards fresh and low-latency content delivery in vehicular networks: An edge caching aspect